Activity
From 2012-01-19 to 2012-02-17
2012-02-16
-
07:33 PM Bug #2115 (Feedback): md5 assembly doesn't compile with gas 2.22
- This updates assembly around the lea instruction to use signed offsets.
It also alters the mboot.S to use non-... -
01:07 AM Bug #1983: sshd problems
- I can confirm this is happening to me too, immediately after upgrading to 151a2. My other zones, which are still on 1...
-
12:35 AM Bug #2087: sd needs to detect and handle SBC-3 correctly
- Current SBC-2 specs also removed these mode pages.
2012-02-15
-
-
08:29 PM Bug #521: syscall number clash with OpenAFS
- I'd like to go ahead and do this. I'm going to go ahead and assign it to myself.
-
05:23 PM Bug #1494 (Resolved): rtls never sets rtlsp->ifname
- Resolved in r13607 commit:bbb551159b47
-
- Resolved in r13606 commit:3720d427d9fa
-
06:10 AM Bug #2131 (Resolved): beadm list -a missing units designation
- Didn't notice this before, but after upgrading OI to
oi_151a2 32-bit (illumos fc320b2833d3)
came across the fol... -
12:26 AM Bug #2130 (Resolved): zvol DKIOCFREE uses nested DMU transactions
- I ran into an issue with COMSTAR's UNMAP handling. COMSTAR uses zvol DKIOCFREE for UNMAP and its involvement ends th...
2012-02-14
-
10:33 PM Bug #2129 (Rejected): interrupt increase on some Sandy Bridge CPUs
- This problem occurs on 151a and with illumos sources downloaded Thursday Feb 9 (compiled with gcc).
The symptoms o... -
- Because ashift is only used in the DMU allocation path, it's probably a sane to permit a smaller ashift in the label...
-
- Typical stack, courtesy of darrenr:
3409: /usr/bin/openssl x509 -noout -hash -in CA-certs/IGC_A.pem
fed335e5 conn... -
- If you need to import a FreeBSD ZFS pool, do not follow the advice in the first comment. Do:
_gpart <device>_ # to e... -
04:32 PM Bug #2127 (In Progress): 'zpool create' core dumps
- I'm unable to reproduce this, but my colleague got this on oi_151a:...
-
- Resolved in r13605 commit:b5c2b5db80d6
-
- changeset: 13604:dfe49af80c1b
tag: tip
user: Bruce Yao <yffbrave@gmail.com>
date: Mon Feb ...
2012-02-13
-
07:25 PM Feature #2124 (Closed): find(1) should support -iname option
- You're right. Trisk told me to make a bug for it, and then it turns out he has approved the patch which adds it, tsk ...
-
07:19 PM Feature #2124: find(1) should support -iname option
- Jeppe Toustrup wrote:
> The @-iname@ option searches for names, non-case sensitive, and is available in GNU find.
... -
- The @-iname@ option searches for names, non-case sensitive, and is available in GNU find.
-
01:39 PM Feature #2123 (New): port dsadm/vmadm from SmartOS
- As a system administrator, I would like Illumos and its children to benefit from the work done in Joyent's SmartOS to...
-
- just noticed this remark (from http://wiki.openindiana.org/display/oi/Workstations+and+Desktops?focusedCommentId=2244...
2012-02-12
-
07:58 PM Bug #1718: MacOS X Lion (10.7) Finder access denied
- Adam K-F wrote:
> Can we get a "target_release" tag installed here? I'm curious to see whether this patch will be i... -
04:19 PM Feature #2100: update HAL to 0.5.14
- Even if we are closer to 0.5.12, we are still setting 0.5.8:
http://src.illumos.org/source/xref/illumos-gate/usr/s... -
02:58 PM Feature #2100: update HAL to 0.5.14
- Our version of HAL is actually closer to 0.5.12 in my experience (I think it has had a few things backported) and in ...
-
- As things stand we treat binutils much like GCC and expect you to be using a known good version. I would not be surp...
2012-02-11
-
- ...
-
02:29 PM Bug #1718: MacOS X Lion (10.7) Finder access denied
- Can we get a "target_release" tag installed here? I'm curious to see whether this patch will be included in 151 stable?
-
- Please, please do it!
-
06:50 AM Bug #2111 (Resolved): begone, pkcs11_kms!
- We have code whose sole reason for being appears to be engaging in congress with a then-StorageTek (now Oracle) tape ...
-
- Yet another bug where someone casts a pointer to a larger sized integer and risks sign extension!
Are we having fun ...
2012-02-10
-
09:27 PM Bug #1983: sshd problems
- Could you provide some further indication of where in these operations you saw delay? Also, could you provide the cor...
-
- This is an issue for the illumos-gate upstream.
-
- inquiry.h: The RDF_ macros actually refer to the SCSI version (SPC version) as found in inq_ansi. The descriptions ar...
-
04:03 PM Bug #2103 (New): dlmgmtd core dump
- I did something that cause @dlmgmtd@ to core dump:...
-
03:42 PM Bug #2083: stmf: worker manager doesn't scale workers anymore
- Dan McDonald wrote:
> Alexander Stetsenko wrote:
> > I guess it should look like this:
>
> Yeah, that's more lik... -
02:08 PM Bug #2102 (New): Resilver restarts if second drive dies
- Hi all
I have a server with four RAIDz2 VDEVs, and the other day, a drive, c14t19d0, died. The spare, c9t7d0, step... -
- The ZFS guys had problems with the Joyent implementation, and I've not heard anything beyond that in months and month...
-
- Resolved in r13602 commit:be505a6a00b0
-
- Resolved in r13603 commit:19698f035619
2012-02-09
-
- Currently Illumos integrates HAL based on version 0.5.8, the last version is 0.5.14. Some newer software (e.g. pulsea...
-
- powertop can freeze during heavy load. For details see:
http://mail.opensolaris.org/pipermail/pm-discuss/2010-Augu... -
08:21 PM Bug #1295: krb5-config only understands krb5
- Here's a possible patch.
-
07:34 PM Bug #1588: nfs4 mirror mount hang
- It's a little bit difficult to provoke this bug but here is what I did to reproduce this bug:
I used two openindia... -
- GNU libc 2.3.4 added RTLD_DEEPBIND:
Place the lookup scope of the symbols in this library ahead of the global scop... -
06:53 PM Feature #2096: Link Layer Discovery protocol support
- I have some preliminary code from the OpenSolaris days that can be used (since I wrote it). However, I do not have a...
-
- LLDP (IEEE 802.1AB) support already in base operation systems (include Solaris 11). Would be much useful to have it i...
-
06:07 PM Bug #2083: stmf: worker manager doesn't scale workers anymore
- Emmanuel Hocdet wrote:
<snip!>
> > Also, could you test the supplied patch stmf.diff, does it really help? (tes... -
- cmlb is aware of this information through DK_TG_GETATTRIBUTE, and assigns a particular number of heads and cylinders.
2012-02-08
-
- Illumos contains private fork of wu-ftpd as its ftpd server. wu-ftpd upstream is dead for 10 years and wu-ftpd was co...
-
- Illumos contains very old rdist implementation, which is protocol-incompatible with "new" 20 years old rdist and that...
-
- There are too many design issues with PCFS so new implementation would be great.
-
06:18 PM Bug #2092 (New): terrible performance from PCFS
- PCFS serializes every operation, synchronously. As a result it's almost certainly the worst performing implementation...
-
- This bug covers only break; unreachable after goto/continue/return/exit in the same function. It is NOT solving cross...
-
- Can you confirm here in the bug's comments that the only removals are for gotos and returns?
-
05:10 PM Bug #1927: nfs4 stale clientid loop (OI server, Linux client)
- Hm spoke too fast. Was able to reproduce with filebench.
-
- If the "pm-capable" property was set by the HBA driver, @sd_pm_setup@ tests START STOP UNIT, POWER CONDITION, and LOG...
-
- The firmware in some newer drives (2TB+) apparently have limited power management support. sd attempts to use power m...
-
- > Emmanuel, could you provide more detailed description of the problem, what are the symptoms… ?
stmf_worker_mgmt(... -
- zdb(1M) currently directs you to not use the utility, unless Sun service tells you to, and how to. That's obviously ...
-
- sd currently issues MODE SENSE for page codes 3 and 4 which have been dropped from SBC-3 (the T10 standard for direct...
-
- This was in the zones SMF log file: ...
2012-02-07
-
- ...
-
10:12 PM Bug #2083: stmf: worker manager doesn't scale workers anymore
- Emmanuel Hocdet wrote:
> it's a regression introduced in changeset: 13291:d5c7e34a4e3f
> (https://www.illumos.org/i... -
- Alexander Stetsenko wrote:
> I guess it should look like this:
Yeah, that's more like it. Dumb question: Will t... -
- I guess it should look like this:
-
- Right idea, wrong way to fix it.
Statics in an MT-environment are dangerous. I see that these variables need to b... -
- Fixed in:
changeset: 13599:b51e5a82e5e2
tag: tip
user: Robert Mustacchi <rm@joyent.com>
date: ... -
12:30 PM Bug #2075: man outputs "geqn should have been given a `-Tutf8' option"
- 'man --recode UTF-8' should be possible in 'man'
it is option from GNU man and is using often for encoding to UTF-8
2012-02-06
-
11:04 PM Bug #2084 (Closed): nfs3_map() should call lm_safemap()
- Sorry, the bug is irrelevant when closed source lock manager is used.
-
- I think that mmap() function of NFSv3 client contains a logical error closely related to nfs3_frlock() function.
A... -
- The rfs4_state_t objects are not released until the client is in the grace period, and the expiration is prevented if...
-
- Test results from Nexenta volunteers (both ET2 and I350) have indicated positive results. I suspect Joyent is using ...
-
- Bruce Yao sent me code for this which was reviewed a long time ago -- here's the latest version of it:
http://mexi... -
- it's a regression introduced in changeset: 13291:d5c7e34a4e3f
(https://www.illumos.org/issues/434)
in the new fun...
2012-02-05
-
- Fixed in:
changeset: 13597:3eac1e8e0f4c
tag: tip
user: Garrett D'Amore <garrett@damore.org>
... -
- I guess I was unclear, let me try to be more clear.
The fact that code is unreachable is easy to determine visuall... -
- I am pretty sure that code like:
return (0);
break;
means that break is unreachable in C :-) -
- You have to be careful here to make sure that the intelligence of the lint you are using, the lint everyone else is u...
-
- the CPR code contains code in cpr_wakecode.s which is much like the CPU boot code in general. Unfortunately, because...
-
06:34 AM Feature #2079 (New): format could support new DKC_BLKDEV
- So, that format output would present MODEL and SERIAL of blkdev device.
-
- changeset: 13596:51d528ff7e15
tag: tip
user: Dan Kruchinin <dan.kruchinin@nexenta.com>
date: ... -
- Oh, and yes, what Albert said. No -dev packages.
-
- For extra clarity, while system/library _could_, it won't. The right way to do that is to move content _out_ of syst...
2012-02-04
-
- ...
-
07:29 PM Feature #2078 (Closed): Support for RFC 4025 and RFC 4322 - using DNSSEC to hold IPsec config
- RFC 4025 and RFC 4322 describe how to use DNSSEC to hold information to relating to the configuration of IPsec Phase ...
-
- Igor S.Kozhukhov wrote:
> I'll be ready with LIBM on next week with checks on SPARC.
> need check full builds with ... -
- I'll be ready with LIBM on next week with checks on SPARC.
need check full builds with lint.
another idea - maybe w... -
11:49 AM Bug #2054: e1000g kstat structure contains needless fields
- http://cr.illumos.org/view/fefwdyvb/e1000g-kstats-03/
-
- The source code contains many unreachable breaks on different places. Unreachable code can be misleading and it is be...
-
- Resolved in r13594 commit:ad255bc59fdf
-
- Resolved in r13593 commit:8bbc95af7713
-
- Resolved in r13592 commit:d4e1700ca091
-
-
- Retargetting for the newest version is fine. :-)
-
- Resolved in:
changeset: 13591:0b83d70b9fe5
tag: tip
user: Garrett D'Amore <garrett@damore.org... -
- This was resolved by the following, I believe:
changeset: 13591:0b83d70b9fe5
tag: tip
user: Gar... -
01:00 AM Bug #2076 (Resolved): blkdev can fail successful xfer if DMA is not used
- This is due to uninitialized variable in bd_xfer_done()
-
- Fixed in:
changeset: 13591:0b83d70b9fe5
tag: tip
user: Garrett D'Amore <garrett@damore.org>
... -
- If $PATH has /usr/gnu/bin before /usr/bin, man(1) will spit out a warning, and happily misrender the man-page. Techn...
-
12:04 AM Bug #2074 (New): dtrace error: translator has already been declared
- Save the following as /var/tmp/tmp/foo.d:...
2012-02-03
-
- Yeah, when libm integrates I want the math headers in system/header properly.
-
- If the AST headers were split off, this becomes a different story... system/header will have no technical reason to d...
-
- Because pkg:/system/header includes the AST headers there is one reference to math.h there:
usr/include/ast/ast_floa... -
- _get_hostserv_inetnetdir_byname() contains FREE_return() which is not nice coding style and is in conflict with the ...
-
- Huawei E173 usb 3g needs usbmodeswitch for switching from USB mass storage to modem supported by usbsacm.
-
06:05 PM Bug #918: Need better IP fanout (esp. with VLANs present)
- Original analysis from Brendan Gregg on this issue follows.
Summary: For high network load (especially 10 GbE on ixg... -
05:01 PM Bug #2071 (Resolved): "libhal_device_free_changeset" function mem-leeks.
- http://src.illumos.org/source/s?refs=libhal_device_free_changeset&project=illumos-gate
"LibHalChangeSetElement" st... -
04:01 PM Bug #1494: rtls never sets rtlsp->ifname
- It seems to me that rtlsp->ifname is not needed at all. We should probably use mac_name(rtlsp->mh) instead. I've post...
-
- This feature is already implemented but not documented correctly in the manual page.
http://src.illumos.org/source... -
- Resolved in r13590 commit:48985d630b76
-
- ...
-
- Well, let's keep it in brand scripts
-
- > I guess zoneadm should check this before calling brand script
I don't think so. The message is from the zone br... -
- ...
-
08:20 AM Bug #2066: out of bounds write in dnode.c
- I agree to Wenchun Cheng, please close this issue, we are going to investigate more deeply.
-
05:59 AM Bug #2066: out of bounds write in dnode.c
- ...
-
- If you're wanting review of the webrev so we can get this integrated, it's more likely to be noticed if emailed to th...
-
- An uninitialized 'err' variable introduced by #1647 means that, depending on the stack layout, zfs(1M) may be unable ...
-
- Resolved in r13589 commit:29b4b175ee0f
2012-02-02
-
- Webrev: http://cr.illumos.org/view/8o1f2dog
-
- In *usr/src/uts/common/fs/zfs/dnode.c*, the OpenSolaris changeset 12684 resulted in the following code:...
-
- Sometimes plockstat is not able to watch hold events (options -H or -A):...
-
- Further removed Jabber and Total_Octets_*. These are jabber_errors and rbytes/obytes now.
-
- A user with SmartOS reported the following panic:...
-
04:36 PM Bug #2051: failed to boot up with Sandy Bridge
- Stephen.
Glad you got your Sandy Bridge with a C600 (Patsburg) chipset up, I run a C204 (Cougar Point) chipset on ... -
04:38 AM Bug #2051: failed to boot up with Sandy Bridge
- I executed the workaround you mentioned about debugger at the grub line. Then turn off apix module. The system can bo...
-
- motivations:
all perl modules on system have style:
library/perl-5/<module name>-<perl version>
will be better t... -
- I'm sick of waiting for a valid build of mps from OI, and wading through build noise.
We should exclude the noise. -
- After feedback from richlowe that the ability of pkgdepend to resolve symlinks in interpreter spec may not extend to ...
-
- the SunOS4 binary compatibility uname intentionally overflows the nodename into the nodeext field (an extension of th...
-
- No amount of logic and thought I can apply suggests why the uint32_t cast is there.
All I can come up with is an a... -
03:42 AM Bug #2060 (New): tmpfs can have distinct files with the same inode number
- http://src.illumos.org/source/xref/illumos-gate/usr/src/uts/common/fs/tmpfs/tmp_tnode.c#172...
-
- Resolved in:
changeset: 13588:4df486ad9a31
tag: tip
user: Theo Schlossnagle <jesus@omniti.com... -
- Various pieces of the kernel contain homebrew (mostly inlined) implementations of offsetof() which cast down to 32bit...
-
- Resolved in r13587 commit:f7bc195459a4
-
- Presently you can direct @nightly(1)@ to send its mail to an arbitrary e-mail address by setting the @MAILTO@ environ...
-
- During other work, I was reminded that meem put together something quite wonderful for cases like this:
http://blo...
2012-02-01
-
- Thanks for the quick reply. I will get the stack trace properly.
The live CD is ok and I actually install 151a to... -
- Hi Stephan,
Please post the start of your stack trace (otherwise no one knows what your issue is). *So I am just ... -
- We are using OpenIndiana build 151a. The hardware has Sandy Bridge CPU.
The system fails to boot after the splash s... -
- Resolved in r13586 commit:57706b7506c8
-
- Resolved in r13586 commit:57706b7506c8
-
- Current behavior:...
-
- The default libumem allocator is the 'instant-fit' allocator. For various reasons, this can cause pathological memory...
-
- At this moment MAC address of the physical interface is visible only after plumbing the link.
It's not possible howe... -
- Also ref to :
...
/platform/platform-name/ufsboot
Second-level program to boot from a disk, DVD, or CD
... -
- man boot tell us:
...
To be able to boot both 32-bit and 64-bit kernels, add
entries for both ker... -
-
- the libssh lint library is built with -l flags which are, in general, reserved for actually _checking_ something with...
-
- Other instances of this message are different (often real) bugs.
Garrett, can we break the circular dependence here? -
- Could we at least put this warning into #ifdef DEBUG?
This story has been going on already 4 years... -
- e1000g kstat structure (e1000g_ksp) includes statistics that are already accounted for by the generic mac statistics ...
-
- Whoever does this could also be awesome, and teach it to valid that the author matches the format we require ("Name n...
-
- Several people have been bitten lately by "recommit" resetting authorship unless told not to.
We can make it bette... -
- So Theo's fix is good.
But your other concerns are wrong. My enhancements compress the strxfrm'd string quite nic... -
-
- This change came from Yuri Pankov, and I forgot to edit the user when I did hg reci. Sorry!
-
- Fixed in:...
-
- iiadm(1M) says:...
2012-01-31
-
- Yuri: I will look into it tonight. (I am about to leave right now, or I would handle it now.)
-
11:03 PM Bug #1332: ssh should use YESEXPR/NOEXPR
- My RTI for this was accepted, but not commited.
-
- I'll take this... its locale related, and I guess this has become kind of one of my areas...
-
11:14 PM Bug #1618: zfs causing system to hang in vmem_xalloc()
- More details as requested by gusev.vitaliy@nexenta.com:...
-
- New locale data is indeed needed, but I think the problem is a root bug in localedef itself. Yao found some of these...
-
- Since this is my buggy code, I'm going to take it.
-
- *Description*
todpc_rtcget() function (see usr/src/uts/i86pc/todpc_subr.c) can return error sometimes if machine is ... -
- Thanks for volunteering. :)
-
- If env is passed a command (_utility_ in the manpage) which it is unable to execute, it currently prints unhelpful an...
-
05:58 PM Feature #2012: ZDB improvements for error-location (with greater verbosity)
- I looked into the code (usr/src/cmd/zdb/zdb.c) and found that sprintf_blkptr_compact() includes many (but not all) of...
2012-01-30
-
- _vt_miocdata_ is missing cases for some VT ioctls including VT_GET_CONSUSER, causing them to always return ENXIO.
-
- The "console owner" is a user logged in on a local console who has exclusive access to local physical resources like ...
-
08:11 PM Bug #1928: zfs send/receive kills COMSTAR
- Haven't tried that. Will initiate a backup tonight and try. Thanks.
-
- Have you tried running "lockstat sleep 10" (or some longer snapshot of time) during the time it's hanging? I don't u...
-
06:42 PM Bug #521: syscall number clash with OpenAFS
- Albert Lee wrote:
> Just in case it matters in the future, it would be good to know whether all of the *at vectors w... -
- $ eject cdrom
eject of cdrom /dev/dsk/c2t1d0s2 failed: hal-storage-can-unmount-volumes-mounted-by-others refused uid... -
11:36 AM Feature #2034: Rewrite perl kstat in c
- David Höppner wrote:
> > > it needs usr/src/common/list/list.c to compile:
> > > @cc kstat.c list.c -lgen -lkstat@
... -
- > > it needs usr/src/common/list/list.c to compile:
> > @cc kstat.c list.c -lgen -lkstat@
>
> Is libgen really ne... -
- David Höppner wrote:
> it needs usr/src/common/list/list.c to compile:
> @cc kstat.c list.c -lgen -lkstat@
Is ... -
- Thanks! No idea, how I've missed it.
I'll just move it to sysmacros.h. And I guess I'll take ARRAY_SIZE along with... -
- There's an implementation in the IB code, here: usr/src/uts/common/sys/ib/clients/rdsv3/rdsv3_impl.h#123
You could... -
- Q: Do we need to add the FreeBSD copyright/license for such trivial piece of code?
-
- The code can be taken from FreeBSD:
#define container_of(p, stype, field) \ ... -
- Resolved in r13584 commit:fc320b2833d3
-
- Resolved in r13583 commit:fabd461f883f
-
- Resolved in r13582 commit:64b0427edfc3
2012-01-29
-
- Missed the question: You can't, because of the whacky \n behaviour.
If I were doing it, I'd probably have impleme... -
- Oh, well, dammit. I just have an _ancient_ /bin/tail for some reason.
Ignore me. -
- GNU tail apparently does 'tail -n1 foo'.
It'd help portability if this feature was not XPG4 specific.
2012-01-28
-
- blkdev calculates bd->d_maxxfer parameter based on the user's dma attrs, or hard-codes it to 1M in case DMA is not us...
-
- This is a "tear down" race involving nsmb_close.
The crash happens in a thread calling close:... -
- After SMB server (domain member) is idle for some time,
we see a panic in the nsmb driver's close function.
Here'... -
- Resolved in r13581 commit:eab0c2354e82
-
- Resolved in r13580 commit:e7b96961e15f
2012-01-27
-
11:04 PM Bug #1970: Boot fails from an rpool on Hitachi 4k sector drives
- zdb produces the following:
MOS Configuration:
version: 28
name: 'rpool'
state: 0
... -
- Information from IRC suggests that zdb of the pool on a Solaris system (such that it can be read), shows the vdevs wi...
-
- One theory here is that moving to 4K drives via resilver has meant that we wrote the data with the old ashift but to ...
-
- I need to do clean up step-by-step here and as I would like to use the latest greatest lint in my CBE, lint reports h...
-
- If you're fixing bugs like this you should:
Edit usr/src/tools/cw/cw.c and remove the forced -Wno-implicit-funct... -
- In many files in kernel there are function declarations for functions defined in different files but the declarations...
-
- Unfortunately thresholds do not have any flags like "SELECT" for elements.
Elements with non-zero "select" are proc... -
- A little excerpt from SES2 specification:
A SELECT bit set to one specifies that the enclosure services process shou... -
- This is a combo fix. Robert Mustacchi ported I350 from FreeBSD for SmartOS, and I (danmcd) added a sprinkling more t...
-
- I reproduced it with the following commands:...
-
- Resolved in r13579 commit:eb88ab5c1435
-
- As above, duplicate of #393
2012-01-26
-
- Resolved in r13577 commit:39608e1faa5f
-
- Resolved in r13578 commit:dbfc0ca2987b
-
- I've taken a short look at this problem and have some comments. I have to admit I'm completely new to Illumos so prob...
-
- Seeing the same here, just got the the assertion failure related to nvidia-driver on ESXi guest..
-
- One aspect of #1779 that I've encountered this being triggered by the nvidia driver, on my systems which have no NVID...
-
- /bin/sleep is the ksh93 sleep
-
- I wonder if a simple thing to do is just parse both , and . the same way here. Perhaps just looking for ispunct. Th...
-
- arbel is really really old infiniband stuff. I agree it needs to be removed.
-
- As the trademark holder for illumos -- it is definitely "illumos" not "Illumos". The reason for this has to do with ...
-
- Resolved in:
changeset: 13576:fcde6b8bbfd2
tag: tip
user: Garrett D'Amore <garrett@damore.org... -
- I think that we actually return "native" for any non-emulating brand. Regardless of the actual name of the brand.
2012-01-25
-
- While working on #2023 I was being very careful because of a complicated amount of unusual (and really crazy) interpo...
-
- At present, we appear to define the native brand in a bunch of places that need special knowledge. Various of them i...
-
- Alexander Eremin wrote:
> This is duplicates of #749, anyway - thanks David, going to test it!
sorry, i did searc... -
- This is duplicates of #749, anyway - thanks David, going to test it!
-
- https://github.com/0xffea/illumos-gate/blob/kstat-rewrite/usr/src/cmd/kstat/kstat.c
-
12:51 PM Bug #968: fct driver sets incorrect fc-ct revision
- Another test with the same initiator-target combination but this time with a McDATA Sphereon 4500, M-EOSc 9.9.9: No e...
-
06:03 AM Bug #968: fct driver sets incorrect fc-ct revision
- Using revision 0x01 on FOS 5.1 and 6.2 (200E) (2009F16) show no signs of regression, the Brocade FOS logs are clean. ...
-
- I implemented a better error handling in nfs4_stub_vnops.c (see attachment). The function nfs4_trigger_domount_args_c...
-
- ipadm has a check for non-native brands, it performs it by comparing the brand string to "native". The native brand ...
-
- bdha reported a bug in some smartos stuff I did whereby networking was busted. This is ultimately because ipadm is r...
2012-01-24
-
- Generic idea is simple: buffer writes on few devices for a long time (maybe hours or days on low-traffic hosts) inste...
-
- Resolved in r13575 commit:36d25dce128e
-
- I'm happy to see the prefetch code remain provided there is a use case for it. The situations I had seen had the pre...
-
- Mat Simon wrote:
> Dan has given instructions to change the value on runtime with:
> echo "fct_create_solct+0x5e2/W... -
- Dan has given instructions to change the value on runtime with:
echo "fct_create_solct+0x5e2/W0x100c641" | mdb -kw ... -
- Chipset is AMD A75 FCH (Hudson D3). I will check PCI IDs in the next days.
Missing is the whole driver for USB3, l... -
- What chipset do you need? Could you provide PCI IDs for your device(s)?
-
- ipf(1M) reference ipf(4) to document the ipfilter config file fornmat, but there's neither such a man page or source ...
-
- Clarifications:
1) for the first "dedup=on" - the error was previously known (found by scrub), and still the block w... -
- While testing interactions of dedup with corrupt on-disk data on my system during repairs, I found that oi_148a (Live...
-
- Tested that when "dedup=verify", and an on-disk block was corrupt (not matching the stored checksum), and the origina...
-
04:02 AM Bug #1954 (Resolved): leak in ZFS from metaslab_group_create and zfs_ereport_checksum
- changeset: 13574:d0fde6cacaac
tag: tip
user: George Wilson <george.wilson@delphix.com>
date: ... -
- changeset: 13574:d0fde6cacaac
tag: tip
user: George Wilson <george.wilson@delphix.com>
date: ... -
- changeset: 13574:d0fde6cacaac
tag: tip
user: George Wilson <george.wilson@delphix.com>
date: ... -
- changeset: 13573:72c96cded60f
tag: tip
user: George Wilson <george.wilson@delphix.com>
date: ... -
- changeset: 13573:72c96cded60f
tag: tip
user: George Wilson <george.wilson@delphix.com>
date: ... -
- changeset: 13573:72c96cded60f
tag: tip
user: George Wilson <george.wilson@delphix.com>
date: ... -
- changeset: 13572:85c66b89d5f2
tag: tip
user: George Wilson <george.wilson@delphix.com>
date: ... -
- changeset: 13571:a5771a96228c
tag: tip
user: Chris Siden <chris.siden@delphix.com>
date: M... -
- changeset: 13570:3411fd5f1589
tag: tip
user: George Wilson <George.Wilson@delphix.com>
date: ... -
- The sort-of-C++ code in libnisdb uses const inconsistently in a way which newer compilers notice.
It's a pain to f... -
- libnisdb uses == to compare strings with string constants. It's in theory possible that this does, or has, worked, s...
2012-01-23
-
- The terminology wrapped around SSD in this case is pretty hairy (at least to me, who knew nothing about it).
You r... -
- changeset: 13569:954bbadc2ca4
tag: tip
user: Dan Kruchinin <dan.kruchinin@nexenta.com>
date: ...
2012-01-22
-
- I had some unrecoverable corruptions in a raidz2 pool, and my research of on-disk data with ZDB led me to blocks whic...
-
- Over the past months I proposed some ideas in the zfs-discuss list [1], which received some praise and some construct...
-
- Over the past months I proposed some ideas in the zfs-discuss list, which received some praise and some constructive ...
-
-
- The separator argument to makedbm is written into a string in rodata, key_sep, and also written _before_ being bounds...
2012-01-21
-
- I object to removal of the code for everybody based on its inefficiency for somebody. For example, vdev prefetch was ...
-
- PREHISTORY
I have recently got reintroduced to pre-fetching mechanisms in ZFS, and particularly to VDEV prefetchin... -
- It should be stressed that the verification reads must get the data from media (bypassing disks' read caches if possi...
-
- My recent research of my corrupted pool data led me to believe that some writes did not land on disks where requested...
-
- While I was researching my pool's data corruption, possibly hitting several deduped files with one corrupt block, it ...
-
- NOTE: I am not sure if this is or is not implemented already. Perhaps the RFE can be closed quickly ;)
I did not g... -
- It was proposed and seems implemented in Solaris 11 but no other distro, to delay actual reuse of those blocks which ...
-
- My "zpool status" reports a corruption in "<metadata>:<0x0>". From the code in http://src.illumos.org/source/xref/ill...
-
- When traversing an object's block tree with "zdb -ddddddd... pool dataset objnum" it might be useful to receive infor...
-
- http://src.illumos.org/source/xref/illumos-gate/usr/src/cmd/zdb/zdb.c#zdb_read_block
This function provides for "z... -
- http://src.illumos.org/source/xref/illumos-gate/usr/src/cmd/zdb/zdb.c#zdb_read_block
This function provides for "z... -
- Currently when a "zpool scrub" locates an error, it tells us that a certain file number or path was broken - contains...
-
- Resolved in r13568 commit:a53b998f5e1b
2012-01-20
-
- changeset: 13567:3cb321bb3838
tag: tip
user: Milan Jurik <milan.jurik@xylab.cz>
date: Fri ... -
12:42 PM Bug #2000: Wrong timestamp on files written over cifs
- Thanks to WinRAR support, I might have more accurate information on how to reproduce this bug.
From Eugene at winr... -
- Newer GCC requires a newer libdwarf with more robust relocation support, something which we would benefit from in gen...
-
- Turns out the word "preamble" was also spelled wrong. I now also have a webrev for you: (now archived) http://www.ke...
2012-01-19
-
- Summarizing an IRC conversation:
With regular shares there's a two-level hierarchy, a group containing shares, wit... -
- A Nexenta engineer who needs to be placed on here will fix this. The only concern I ever had about this were any bus...
-
- int
md_trans_ioctl(dev_t dev, int cmd, caddr_t data, int mode, IOLOCK *lockp)
It implements md_ioctl which shoul... -
- WORKAROUND:
As was suggested in zfs-discuss list, setting dedup attribute to "verify" instead of "on" did somewhat h... -
- We have a ZFS volume shared over CIFS, accessed by clients on windows computers.
When a user unrar a .rar file wit... -
09:45 AM Bug #393: Invalid nd_hostservlist contents crash mountd
- Another crash:
libc_hwcap1.so.1`_lwp_kill+0x15
libc_hwcap1.so.1`raise+0x22
libc_hwcap1.so.1`abort+0xf2
... -
- Resolved in r13566 commit:b2b4f797c428
Also available in: Atom